ASUS
GPU & AI Servers
ASUS XA NB3I-E12
9U NVIDIA HGX B300 8-GPU server with dual Intel Xeon 6 for heavy AI training and inference.
- GPUs: NVIDIA HGX B300 8-GPU
- CPUs: Dual Intel Xeon 6
- Networking: 8× embedded ConnectX-8 InfiniBand, dual 10 GbE
ASUS ESC8000-E12P
Model ESC8000-E12P
Intel Xeon 6 4U NVIDIA MGX server for eight dual-slot GPUs or Gaudi 3 accelerators.
- GPUs8× dual-slot: H200, RTX PRO 6000 Blackwell SE or Intel Gaudi 3
- CPUsDual Intel Xeon 6
- MemoryUp to 32 DIMM
- Expansion5× PCIe 5.0
Overview
About the ESC8000-E12P
The ASUS ESC8000-E12P is a high-density 4U server with dual Intel Xeon 6 processors and room for eight dual-slot GPUs, active or passive. It is built on the NVIDIA MGX modular architecture.
Broadcom PCIe 5.0 switches give every GPU full x16 bandwidth, and the tool-less chassis can be completely disassembled without tools for fast servicing.
Eight dual-slot GPUs
NVIDIA H200 NVL, RTX PRO 6000 Blackwell or Intel Gaudi 3.
PCIe 5.0 switching
Broadcom PEX89000 switches with 128 GB/s per x16 port.
Intel Xeon 6
All P-core processors with DDR5 memory.
NVIDIA MGX
Up to 160 configurations to tailor the system.
Tool-less maintenance
Fully disassembles without tools.
Management and security
ASMB12-iKVM, ASUS Control Center and TPM 2.0.
Specifications
Technical Specifications
Processor & memory
| CPUs | 2× LGA 4710, Intel Xeon 6500 / 6700, up to 350 W |
|---|---|
| Memory slots | 32 (8 channels per CPU, 16 DIMMs per CPU) |
| Memory capacity | Up to 4 TB |
| Memory type | DDR5-6400 (1DPC) / 5200 (2DPC) RDIMM / 3DS RDIMM |
GPU & expansion
| GPU slots | 8× PCIe x16 Gen5 for dual-slot GPUs (FHFL) |
|---|---|
| NIC / DPU slots | 5× PCIe x16 Gen5 (FHFL) for NIC or BlueField-3 |
| Additional | 1× PCIe x16 (Gen5 x8, FHHL) |
Storage & networking
| Storage | 8× 2.5" front hot-swap (up to 8 NVMe); 2× M.2 (Gen5 x4, up to 22110) |
|---|---|
| Networking | 2× 10 GbE (X710-AT2), 1× management |
Chassis & power
| Form factor | 4U, NVIDIA MGX |
|---|---|
| Dimensions | 800 × 439.5 × 175 mm |
| Weight | 45 kg net / 56 kg gross |
| Power | 3+1 redundant 3200 W 80 PLUS Titanium |
| Operating temperature | 10 °C to 35 °C |
Source: official ASUS product information.
Specifications are set by the manufacturer and may change. Plan Man confirms the final configuration, availability and lead time in your quote.
Use cases
What it's built for
- Enterprise AI training and inference
- Machine learning
- HPC
- GPU virtualisation
Plan Man services
Supplied and supported in the UAE
- Requirement analysis, sizing and bill of materials
- Genuine supply with manufacturer warranty
- Rack integration, cabling and onsite deployment
- Power and liquid cooling planning for dense GPU racks
- After-sales support, RMA handling and upgrades from Dubai
ASUSNew
GPU & AI Servers
ASUS ESC8000A-E13X
4U NVIDIA RTX PRO Server with ConnectX-8 SuperNICs and eight RTX PRO 6000 Blackwell GPUs.
- GPUs: 8× RTX PRO 6000 Blackwell Server Edition
- CPUs: Dual AMD EPYC 9005
- Networking: 8× 400 Gb/s QSFP (ConnectX-8)
ASUS
GPU & AI Servers
ASUS ESC8000A-E13
AMD EPYC 4U server supporting eight dual-slot NVIDIA H200 or RTX PRO 6000 Blackwell GPUs.
- GPUs: 8× dual-slot: H200, RTX PRO 6000 Blackwell SE, AMD Instinct MI350P
- CPUs: Dual AMD EPYC 9005/9004
- Memory: Up to 24 DIMM
Get a Quote
Tell us your workload and timeline for the ASUS ESC8000-E12P. We reply within one business day with a configuration, pricing and lead time.